Breast cancer is the most common cancer in Hawaii. The age-adjusted incidence rate has increased from 40 per 100,000 women in 1960 to 110 per 100,000 women in 1990. During the same time availability of mammography screening (MU) has increased greatly. The goal of this project is to estimate what part of the increase in breast cancer incidence rates is attributable to increased MU. This epidemiologic study will use an innovative research methodology linking insurance claims data, cancer registry data, census information, and vital records data on a population based level. The hypothesis that geographic areas with high MU experience both higher breast cancer incidence rates and diagnosis at earlier stages than geographic areas with low MU will be tested on a ZIP code level. The study population consists of all civilian women 40 years and over, who resided in Hawaii between 1990 and 1992, approximately 200,000 women. The majority of Hawaii's population are of Asian and Pacific Islander ancestry. Data on MU will be obtained from all health plans in the state. Nearly 70% of the population are already part of the Hawaii Insurance Claims Surveillance Project (HICSP). MU data for the rest will be purchased from the health plans not included in HICSP. Breast cancer incidence data for 1992 will be obtained from the Hawaii Tumor Registry, part of the SEER program. Data will be analyzed using multiple logistic regression adjusting for age, ethnicity, tumor stage, and age at first birth. In addition, a case-control study will compare the impact of MU on tumor stage at diagnosis in women with late stage vs. early stage disease. Women with breast cancer who are part of HICSP will be included. Linking of data bases will occur with soundex coded names. Relative risk estimates (odds ratios) of being diagnosed with early vs. late stage disease will be calculated. Noteworthy features of this project are the population based approach using different data bases, the opportunity to study a closed geographic area, and the ethnic variety of the study subjects.